The goal of this project is to identify genes critical to the development of breast or ovarian cancer. Cancer genes have often been revealed in one of two ways: as genes whose expression is regulated by known tumor suppressors or as genes with function-abrogating mutations in tumors and families. This project proposes to integrate these two approaches, by exploiting two functions of BRCA1: tumor suppression and transcriptional regulation. We have developed breast, ovarian, and other epithelial cell lines in which BRCA1 expression can be experimentally controlled. Expression profiles of cells in which over-expression of BRCA1 has been induced will be compared to profiles of uninduced cells using arrayed cDNA sequences. This comparison will allow identification of genes whose expression is influenced by wild type BRCA1 but not by mutant BRCA1. Differentially expressed genes ("candidate genes") identified in cell lines will be tested in several ways for relevance to human tumorigenesis. Expression levels of candidate genes will be compared in breast and ovarian tumors vs. normal epithelial cells from the same patients. Breast and ovarian tissues are available from patients with sporadic cancer and from patients with inherited cancer due to known BRCA1 or BRCA2 mutations. Candidate genes will be screened for somatic mutations in these tumors. Genes altered in primary tumors may represent critical steps in sporadic carcinogenesis and/or modifiers of the effects of BRCA1 (or BRCA2) in inherited disease. Concurrently, families in our series with inherited breast cancer not attributable to BRCA1 or BRCA2 will be evaluated for linkage of cancer predisposition to other chromosomal regions. Constitutional DNA from members of these families will be tested for germline mutations in candidate genes that map to potential regions of linkage. BRCA1-regulated genes of special interest would be of several classes: any genes that are expressed only in breast and ovary, possibly explaining the tumor specificity of BRCA1; any genes over-expressed on the surface of tumor cells, possibly permitting exploitation as a diagnostic or as a therapeutic; or any genes over-expressed in tumors with a product secreted early in tumor development, possibly yielding an early diagnostic marker. The search for such genes has been a central problem of breast cancer biology. The intent of this project is to exploit genomic technologies to integrate the identification of genes biologically related to BRCA1 with characterization of those genes in human tumors and families.